o?i the Direction of Evolution. 179


primary bars, or adult or secondary bars. The fact of derivation

remains the same. If we find cases in which the deletion of spots

has been carried still further, until only a few of the larger spots

in the posterior row remain, as in Zenaida and Zenaidura, we

may then say that ' bars ' have preceded spots. But in this case,

the 'bars' are only rows of chequers, and the permanent 'spots'

are the true homologues of typical bar-elements in other species.


I believe that the few spots in the wing of the adult

Calopelia are such homologues, and that they are preceded in the

juvenal plumage by corresponding marks which have been over-

looked. I am aware that in the young of this species, the

scapulars, wing-coverts, and tertials are said to be "barred with

black." These juvenal bars are probably of the same nature as

those seen in the young Tambourine, the relation of which to

chequers has already been indicated.


Although it may now be sufficiently clear that my view

does not stand in contradiction with Dr. Butler's observations on

African doves, it may not be amiss to pursue the subject a little

further.


Specific characters often appear to come in suddenly and

to be immutable. This seems often to be so in passing from the

juvenal to the mature plumage, with one or several moults. The

earlier color-pattern is often succeeded by one so different that

we are puzzled to see any possible way of continuous differentia-

tion from one to the other. Here, if anywhere, we should expect

to see the mutationist triumphant. His biometric curves would

surely declare gaps in abundance beyond the span of any con-

ceivable bridges. Just as surely is he a too-willing believer in

miracles. The supposed breaks in continuity between stages are,

at least in many cases, only discontinuities in observation.


It is the history of specific characters that turns apparent

discontinuities into continuities; the lack of it that multiplies

mutations into premutations, and premutations into hypothetic

pangens. Complete histories are indeed rarely attainable, but so

much the more significant are they, if just so far as they can be

established, they remove all necessity for resort to mutation

hypotheses.


In illustration, I will take the case of the little Diamond



